A method and apparatus for dehazing an image

Abstract

The application discloses a haze-removed image generation method and device, and relates to the technical field of image processing. The target echo time is calculated based on the noise-filtered denoising photon echo data. The target distance is calculated according to the speed of light and the target echo time. The backscattering intensity map is calculated based on the above data and a preset buffer interval. The target scattering coefficient is calculated based on the backscattering intensity map, a first calibration constant and the target distance. The atmospheric transmittance is calculated based on the target scattering coefficient and a preset atmospheric scattering proportion factor. The polarization initial atmospheric transmittance is calculated based on a four-channel polarization intensity map. The target atmospheric transmittance parameter is calculated based on the atmospheric transmittance and the polarization initial atmospheric transmittance. The atmospheric scattering light value is calculated based on the target atmospheric transmittance parameter. The haze-removed image is calculated based on the target image, the target atmospheric transmittance parameter and the atmospheric scattering light value. In this way, the image quality of the haze-removed image is improved.

Description

Technical Field

[0001] This application relates to the field of image processing technology, and in particular to a method and apparatus for generating dehazed images. Background Technology

[0002] Polarization-based fog-penetrating imaging technology utilizes the difference in polarization states between scattered light and target reflected light in a foggy environment to separate the target signal from scattering interference by analyzing polarization information. The core of this type of passive imaging method lies in the accurate estimation of atmospheric transmittance.

[0003] In related technologies, atmospheric transmittance is estimated by inverting the transmittance distribution map based on the changes in polarization parameters. However, in actual complex atmospheric environments, passive polarization observation is easily affected by ambient light interference and insufficient signal-to-noise ratio in weak texture regions, leading to deviations in transmittance estimation and affecting the authenticity and consistency of defogging effects. Summary of the Invention

[0004] In view of this, this application provides a method and apparatus for generating dehazed images.

[0005] The objective of this application can be achieved through the following technical solutions:

[0006] The first aspect of this application is to provide a method for generating dehazed images, including:

[0007] Acquire raw photon echo data and four-channel polarization intensity images;

[0008] The original photon echo data is subjected to noise filtering to obtain denoised photon echo data. The original photon echo data includes the photon count in each time window of each pixel.

[0009] Based on the denoised photon echo data, the target echo time of each pixel is calculated;

[0010] The target distance for each pixel is calculated based on the speed of light and the target echo time.

[0011] Based on the denoised photon echo data, the target echo time of each pixel, and the preset buffer interval, the backscattering intensity map is calculated.

[0012] The target scattering coefficient is calculated based on the backscattering intensity map, the first calibration constant, and the target distance for each pixel;

[0013] The atmospheric transmittance of each pixel is calculated based on the target scattering coefficient and the preset atmospheric scattering scaling factor.

[0014] The initial atmospheric transmittance was calculated based on the four-channel polarization intensity image.

[0015] Based on atmospheric transmittance and initial polarization atmospheric transmittance, the target atmospheric transmittance parameter for each pixel is calculated.

[0016] Based on the target atmospheric transmittance parameter, the atmospheric scattered light value of each pixel is calculated;

[0017] The dehazed image is calculated based on the target image, the target atmospheric transmittance parameter, and the atmospheric scattered light value.

[0018] In one optional embodiment, the initial atmospheric transmittance based on the four-channel polarization intensity image is calculated, including:

[0019] Stokes parameters are calculated based on four-channel polarization intensity images. The Stokes parameters include the first Stokes parameter, the second Stokes parameter, and the third Stokes parameter.

[0020] The initial atmospheric transmittance based on the Stokes parameters was calculated.

[0021] In one optional embodiment, the Stokes parameters are calculated based on the four-channel polarization intensity image, including:

[0022] Based on the four-channel polarization intensity image, the Stokes parameters are calculated using the following formula:

[0023] ;

[0024] in, Indicates the first Stokes parameter. This represents the polarization intensity image in the 0° polarization direction. This represents a polarization intensity image along a 45° polarization direction. This represents a polarization intensity image along a 90° polarization direction. A polarization intensity image representing the polarization direction at 135°;

[0025] ;

[0026] in, Indicates the second Stokes parameter. This represents the polarization intensity image in the 0° polarization direction. A polarization intensity image representing the polarization direction at 90°;

[0027] ;

[0028] in, Indicates the third Stokes parameter. This represents a polarization intensity image along a 45° polarization direction. This image represents the polarization intensity along the 135° polarization direction.

[0029] In one optional embodiment, the target atmospheric transmittance parameter for each pixel is calculated based on atmospheric transmittance and initial polarization atmospheric transmittance, including:

[0030] Based on atmospheric transmittance and initial polarization atmospheric transmittance, the target atmospheric transmittance parameter for each pixel is calculated using the following formula:

[0031] ;

[0032] in, This represents the target atmospheric transmittance parameter of the pixel in the i-th row and j-th column. Indicates adaptive weights, Indicates atmospheric transmittance. This represents the initial atmospheric transmittance due to polarization.

[0033] In an optional embodiment, before calculating the target atmospheric transmittance parameter for each pixel based on atmospheric transmittance and initial polarization atmospheric transmittance, the method further includes:

[0034] The target time window is determined based on the target echo time of each pixel and a preset buffer interval.

[0035] The adaptive weights are calculated using the following formula:

[0036] ;

[0037] in, For adaptive weights, Indicates the first signal-to-noise ratio. Indicates the second signal-to-noise ratio;

[0038] ;

[0039] in, Indicates the first signal-to-noise ratio. Indicates the first calibration coefficient. This represents the maximum photon count within the target time window;

[0040] ;

[0041] in, This indicates the second signal-to-noise ratio. Indicates the second calibration coefficient. Indicates median filtering. It is the gradient operator. Indicates to Find the gradient. This indicates the calculation of the Frobenius norm.

[0042] In one optional embodiment, the target echo time for each pixel is calculated based on the denoised photon echo data, including:

[0043] ;

[0044] in, This represents the target echo time of the pixel in the i-th row and j-th column. This indicates finding the maximum value of the expression within the parentheses. This represents the photon count in the t-th time window of the i-th row and j-th column of the denoised photon echo data. This represents the system's impulse response function.

[0045] In one optional embodiment, a backscattering intensity map is calculated based on denoised photon echo data, the target echo time of each pixel, and a preset buffer interval, including:

[0046] Based on the denoised photon echo data, the target echo time of each pixel, and the preset buffer interval, the backscattering intensity map is calculated using the following formula:

[0047] ;

[0048] in, This represents the backscattering intensity map of the pixel in the i-th row and j-th column. This represents the target echo time of the pixel in the i-th row and j-th column. Indicates the preset buffer interval. This represents the photon count in the t-th time window of the i-th row and j-th column pixel in the denoised photon echo data.

[0049] In one optional embodiment, the target scattering coefficient is calculated based on the backscattering intensity map, a first calibration constant, and the target distance for each pixel, including:

[0050] Based on the backscattering intensity map, the first calibration constant, and the target distance for each pixel, the target scattering coefficient is calculated using the following formula:

[0051] ;

[0052] in, This represents the target scattering coefficient of the i-th row and j-th column pixel. This represents the first calibration constant. This represents the backscattering intensity map of the pixel in the i-th row and j-th column. This represents the target distance of the pixel in the i-th row and j-th column.

[0053] In one optional embodiment, the atmospheric transmittance of each pixel is calculated based on the target scattering coefficient and a preset atmospheric scattering scaling factor, including:

[0054] Based on the target scattering coefficient and the preset atmospheric scattering scaling factor, the atmospheric transmittance of each pixel is calculated using the following formula:

[0055] ;

[0056] in, This represents the atmospheric transmittance of the pixel in the i-th row and j-th column. This represents the target scattering coefficient of the i-th row and j-th column pixel. This represents the preset atmospheric scattering scaling factor. This represents the target distance of the pixel in the i-th row and j-th column.

[0057] A second aspect of this application is to provide a dehazing image generation apparatus, comprising:

[0058] The acquisition module is used to acquire raw photon echo data and four-channel polarization intensity images;

[0059] The filtering module is used to perform noise filtering on the original photon echo data to obtain denoised photon echo data. The original photon echo data includes the photon count in each time window of each pixel.

[0060] The first calculation module is used to calculate the target echo time of each pixel based on the denoised photon echo data;

[0061] The second calculation module is used to calculate the target distance for each pixel based on the speed of light and the target echo time.

[0062] The third calculation module is used to calculate the backscattering intensity map based on the denoised photon echo data, the target echo time of each pixel and the preset buffer interval.

[0063] The fourth calculation module is used to calculate the target scattering coefficient based on the backscattering intensity map, the first calibration constant, and the target distance for each pixel;

[0064] The fifth calculation module is used to calculate the atmospheric transmittance of each pixel based on the target scattering coefficient and the preset atmospheric scattering scaling factor.

[0065] The sixth calculation module is used to calculate the initial atmospheric transmittance based on the four-channel polarization intensity image;

[0066] The seventh calculation module is used to calculate the target atmospheric transmittance parameter for each pixel based on atmospheric transmittance and initial polarization atmospheric transmittance.

[0067] The eighth calculation module is used to calculate the atmospheric scattered light value of each pixel based on the target atmospheric transmittance parameter;

[0068] The ninth calculation module is used to calculate the dehazed image based on the target image, the target atmospheric transmittance parameter, and the atmospheric scattered light value.

[0069] A third aspect of this application is to provide an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to perform the method as described in the first aspect.

[0070] A fourth aspect of this application is to provide a computer-readable storage medium storing a computer program, which, when executed by a processor, performs the method as described in the first aspect.

[0071] Compared with existing technologies, the dehazing image generation method provided in this application calculates the target echo time based on noise-filtered denoised photon echo data; calculates the target distance based on the speed of light and the target echo time; calculates a backscattering intensity map based on the above data and a preset buffer interval; calculates the target scattering coefficient based on the backscattering intensity map, a first calibration constant, and the target distance; calculates the atmospheric transmittance based on the target scattering coefficient and a preset atmospheric scattering scaling factor; calculates the initial polarization atmospheric transmittance based on a four-channel polarization intensity image; calculates the target atmospheric transmittance parameter based on the atmospheric transmittance and the initial polarization atmospheric transmittance; calculates the atmospheric scattered light value based on the target atmospheric transmittance parameter; and calculates the dehazing image based on the target image, the target atmospheric transmittance parameter, and the atmospheric scattered light value. This improves the image quality of the dehazing image. Attached Figure Description

[0072] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0073] Figure 1 This is a schematic flowchart of a dehazing image generation method provided in an embodiment of this application;

[0074] Figure 2 A schematic diagram of a process for acquiring raw photon echo data and a four-channel polarization intensity image based on a laser source and a single-photon detector, as provided in an embodiment of this application;

[0075] Figure 3This is another flowchart illustrating the dehazing image generation method provided in an embodiment of this application;

[0076] Figure 4 A structural block diagram of the dehazing image generation apparatus provided in the embodiments of this application;

[0077] Figure 5 This is a structural block diagram of an electronic device for implementing a dehazing image generation method, provided in an embodiment of this application. Detailed Implementation

[0078] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort should fall within the scope of protection of the present application.

[0079] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0080] It should be understood that in the embodiments of this application, "at least one" means one or more, and "more than one" means two or more. "And / or" is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, and B existing alone. The character " / " generally indicates that the related objects before and after it are in an "or" relationship. "Contains A, B and / or C" means containing any one, two, or three of A, B, and C.

[0081] It should be understood that in the embodiments of this application, "B corresponding to A", "B corresponding to A", "A corresponds to B" or "B corresponds to A" means that B is associated with A, and B can be determined based on A. Determining B based on A does not mean that B is determined solely based on A; B can also be determined based on A and / or other information.

[0082] To address the technical problems existing in related technologies, this application provides a method and apparatus for generating dehazed images.

[0083] The dehazing image generation method provided in this application can be executed by an electronic device, such as a terminal or a server. The terminal can be a smartphone, tablet, laptop, or other similar device. The server can be a standalone physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud storage, network services, cloud communication, middleware services, domain name services, security services, CDN (Content Delivery Network), and big data and artificial intelligence platforms. It is understood that this application does not limit the specific entity executing the dehazing image generation method.

[0084] The technical solution of this application will be described in detail below through specific embodiments. It should be noted that the following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments. The embodiments described below are used to explain the technical solution of this application and are not intended to limit actual use.

[0085] To address the technical problems existing in related technologies, embodiments of this application provide a method for generating dehazed images, such as... Figure 1 As shown, Figure 1 This is a flowchart of a dehazing image generation method provided in an embodiment of this application. It should be noted that the steps shown may be executed in a different logical order than that shown in the flowchart. The method may include the following steps S101 to S111.

[0086] Step S101: Acquire raw photon echo data and four-channel polarization intensity images.

[0087] It should be noted that the raw photon echo data is a three-dimensional matrix, which includes the echo occurrence time and photon count collected by the single-photon detector.

[0088] In one optional embodiment, the four-channel polarization intensity image includes a polarization intensity image in the 0° polarization direction, a polarization intensity image in the 45° polarization direction, a polarization intensity image in the 90° polarization direction, and a polarization intensity image in the 135° polarization direction.

[0089] Figure 2 This application provides a schematic diagram of a process for acquiring raw photon echo data and a four-channel polarization intensity image based on a laser source and a single-photon detector, as shown in the embodiments of this application. Figure 2As shown, a laser source emits laser pulses, which are then emitted onto the imaging target. Upon encountering the target, the laser pulses are reflected, forming an echo signal. This echo signal contains information such as the shape and distance of the imaging target. A synchronization module coordinates the timing between the laser source and the single-photon detector, ensuring precise matching between laser pulse emission and echo signal reception. After receiving the echo signal, the single-photon detector records the arrival time and position of each photon, thus constructing three-dimensional point cloud data of the imaging target—the raw photon echo data. This raw photon echo data reveals the three-dimensional structure of the imaging target. A four-channel polarization intensity image is acquired using a polarization sensor; this image is used to display the polarization characteristics of the imaging target.

[0090] Step S102: Perform noise filtering on the original photon echo data to obtain denoised photon echo data.

[0091] It should be noted that the raw photon echo data includes the photon count for each time window within each pixel.

[0092] In one optional embodiment, noise filtering is performed on the original photon echo data to obtain denoised photon echo data, specifically including the following steps:

[0093] The following formula is used to filter noise from the original photon echo data to obtain denoised photon echo data:

[0094] (1);

[0095] in, This represents the photon count in the t-th time window of the i-th row and j-th column of the denoised photon echo data. This represents the filtering threshold set based on the Poisson distribution. This represents the i-th pixel in the i-th row and j-th column of the original photon echo data. Photon count within a time window.

[0096] By applying noise filtering to the raw photon echo data, denoised photon echo data is obtained, significantly enhancing the contrast between the photon echo signal and noise, making the echo peaks clearer. Furthermore, the filtered echo signal is smoother, and the determined target echo time is more accurate, which is beneficial for high-precision time synchronization.

[0097] Step S103: Calculate the target echo time for each pixel based on the denoised photon echo data.

[0098] In one optional embodiment, the target echo time for each pixel is calculated based on the denoised photon echo data, including:

[0099] (2);

[0100] in, This represents the target echo time of the pixel in the i-th row and j-th column. This indicates finding the maximum value of the expression within the parentheses. This represents the photon count in the t-th time window of the i-th row and j-th column of the denoised photon echo data. This represents the system's impulse response function.

[0101] In one alternative embodiment, It is set according to the actual situation.

[0102] Step S104: Calculate the target distance for each pixel based on the speed of light and the target echo time.

[0103] In one optional embodiment, calculating the target distance for each pixel based on the speed of light and the target echo time includes:

[0104] Based on the speed of light and the target echo time, the target distance for each pixel is calculated using the following formula:

[0105] (3);

[0106] in, This represents the target distance of the i-th row and j-th column pixel. Represents the speed of light. This represents the target echo time of the pixel in the i-th row and j-th column.

[0107] Step S105: Based on the denoised photon echo data, the target echo time of each pixel, and the preset buffer interval, calculate the backscattering intensity map.

[0108] In one optional embodiment, a backscattering intensity map is calculated based on denoised photon echo data, the target echo time of each pixel, and a preset buffer interval, including:

[0109] Based on the denoised photon echo data, the target echo time of each pixel, and the preset buffer interval, the backscattering intensity map is calculated using the following formula:

[0110] (4);

[0111] in, This represents the backscattering intensity map of the pixel in the i-th row and j-th column. This represents the target echo time of the pixel in the i-th row and j-th column. Indicates the preset buffer interval. This represents the photon count in the t-th time window of the i-th row and j-th column pixel in the denoised photon echo data.

[0112] In a more specific embodiment, the preset buffer interval can be set according to the actual situation.

[0113] Step S106: Calculate the target scattering coefficient based on the backscattering intensity map, the first calibration constant, and the target distance for each pixel.

[0114] In one optional embodiment, the target scattering coefficient is calculated based on the backscattering intensity map, a first calibration constant, and the target distance for each pixel, including:

[0115] Based on the backscattering intensity map, the first calibration constant, and the target distance for each pixel, the target scattering coefficient is calculated using the following formula:

[0116] (5);

[0117] in, This represents the target scattering coefficient of the i-th row and j-th column pixel. This represents the first calibration constant. This represents the backscattering intensity map of the pixel in the i-th row and j-th column. This represents the target distance of the pixel in the i-th row and j-th column.

[0118] Step S107: Calculate the atmospheric transmittance of each pixel based on the target scattering coefficient and the preset atmospheric scattering scaling factor.

[0119] In one optional embodiment, the atmospheric transmittance of each pixel is calculated based on the target scattering coefficient and a preset atmospheric scattering scaling factor, including:

[0120] Based on the target scattering coefficient and the preset atmospheric scattering scaling factor, the atmospheric transmittance of each pixel is calculated using the following formula:

[0121] (6);

[0122] in, This represents the atmospheric transmittance of the pixel in the i-th row and j-th column. This represents the target scattering coefficient of the i-th row and j-th column pixel. This represents the preset atmospheric scattering scaling factor. This represents the target distance of the pixel in the i-th row and j-th column.

[0123] Step S108: Calculate the initial atmospheric transmittance based on the four-channel polarization intensity image.

[0124] In one optional embodiment, the initial atmospheric transmittance based on the four-channel polarization intensity image is calculated, including:

[0125] Stokes parameters are calculated based on the four-channel polarization intensity image. The Stokes parameters include the first Stokes parameter, the second Stokes parameter, and the third Stokes parameter. Based on the Stokes parameters, the initial atmospheric transmittance is calculated.

[0126] In one specific embodiment, the Stokes parameters are calculated based on the four-channel polarization intensity image, including:

[0127] Based on the four-channel polarization intensity image, the Stokes parameters are calculated using the following formula:

[0128] (7);

[0129] in, Indicates the first Stokes parameter. This represents the polarization intensity image in the 0° polarization direction. This represents a polarization intensity image along a 45° polarization direction. This represents a polarization intensity image along a 90° polarization direction. A polarization intensity image representing the polarization direction at 135°;

[0130] (8);

[0131] in, Indicates the second Stokes parameter. This represents the polarization intensity image in the 0° polarization direction. A polarization intensity image representing the polarization direction at 90°;

[0132] (9);

[0133] in, Indicates the third Stokes parameter. This represents a polarization intensity image along a 45° polarization direction. This image represents the polarization intensity along the 135° polarization direction.

[0134] In one specific embodiment, the initial atmospheric transmittance based on the Stokes parameters is calculated, including:

[0135] Based on the Stokes parameters, the initial atmospheric transmittance for polarization is calculated using the following formula:

[0136] (10);

[0137] in, Indicates the initial atmospheric transmittance under polarization. Indicates the first Stokes parameter. Indicates the second Stokes parameter. This represents the third Stokes parameter.

[0138] Step S109: Calculate the target atmospheric transmittance parameter for each pixel based on atmospheric transmittance and initial polarization atmospheric transmittance.

[0139] In one optional embodiment, the target atmospheric transmittance parameter for each pixel is calculated based on atmospheric transmittance and initial polarization atmospheric transmittance, including:

[0140] Based on atmospheric transmittance and initial polarization atmospheric transmittance, the target atmospheric transmittance parameter for each pixel is calculated using the following formula:

[0141] (11);

[0142] in, This represents the target atmospheric transmittance parameter of the pixel in the i-th row and j-th column. Indicates adaptive weights, Indicates atmospheric transmittance. This represents the initial atmospheric transmittance due to polarization.

[0143] In another optional embodiment, the dehazing image generation method provided in this application further includes:

[0144] The target time window is determined based on the target echo time of each pixel and a preset buffer interval.

[0145] The adaptive weights are calculated using the following formula:

[0146] (12);

[0147] in, For adaptive weights, Indicates the first signal-to-noise ratio. Indicates the second signal-to-noise ratio;

[0148] (13);

[0149] in, Indicates the first signal-to-noise ratio. Indicates the first calibration coefficient. This represents the maximum photon count within the target time window;

[0150] (14);

[0151] in, This indicates the second signal-to-noise ratio. Indicates the second calibration coefficient. This represents median filtering, where ▽ is the gradient operator. Indicates to Find the gradient. This indicates the calculation of the Frobenius norm.

[0152] Step S110: Calculate the atmospheric scattered light value for each pixel based on the target atmospheric transmittance parameter.

[0153] In one optional embodiment, the atmospheric scattered light value of each pixel is calculated based on the target atmospheric transmittance parameter, including:

[0154] Based on the target atmospheric transmittance parameter, the atmospheric scattered light value for each pixel is calculated using the following formula:

[0155] (15);

[0156] in, This represents the atmospheric scattered light value of the pixel in the i-th row and j-th column. This represents the first Stokes parameter of the pixel in the i-th row and j-th column. This represents the target atmospheric transmittance parameter for the pixel in the i-th row and j-th column.

[0157] Step S111: Calculate the dehazed image based on the target image, the target atmospheric transmittance parameter, and the atmospheric scattered light value.

[0158] In one optional embodiment, a dehazed image is calculated based on the target image, the target atmospheric transmittance parameter, and the atmospheric scattered light value, including:

[0159] (16);

[0160] in, Represents a dehazed image. This represents the atmospheric scattered light value of the pixel in the i-th row and j-th column. This represents the first Stokes parameter of the pixel in the i-th row and j-th column. This represents the target atmospheric transmittance parameter for the pixel in the i-th row and j-th column.

[0161] In another optional embodiment, the dehazing image generation method provided in this application further includes:

[0162] Generate a target distance map based on the target distance;

[0163] Extract edge features from the target distance map;

[0164] The edge features are normalized to obtain a normalized depth edge image; based on the normalized depth edge image, the dehazed image is enhanced to obtain an enhanced dehazed image.

[0165] In a more specific embodiment, generating a target distance map based on the target distance includes the following steps: creating a two-dimensional grid, where each grid cell represents a pixel position; filling the corresponding grid cell with the target distance calculated for each point to form a distance matrix; and visualizing the distance matrix to obtain the target distance map.

[0166] In a more specific embodiment, the normalized depth edge image is obtained using the following formula:

[0167] (17);

[0168] (18);

[0169] in, Represents a normalized depth edge image. Represents the target distance map. This represents the Sobel operator in the horizontal direction. Represents the Sobel operator in the vertical direction. This indicates normalization.

[0170] In another, more specific embodiment, based on the normalized depth edge image, the dehazed image is subjected to detail enhancement processing using the following formula to obtain an enhanced dehazed image:

[0171] (19);

[0172] in, This represents the i-th channel of the enhanced dehazed image. This represents the i-th channel of the dehazed image. This represents a normalized depth edge image.

[0173] This application also provides a flowchart illustrating a dehazing image generation method, such as... Figure 3 As shown, the specific steps include:

[0174] Step 1: Perform noise reduction preprocessing on the raw photon echo data to obtain denoised photon echo data;

[0175] Step 2: Calculate the backscattering intensity map based on the denoised photon echo data;

[0176] Step 3: Calculate the atmospheric transmittance based on the backscattering intensity map;

[0177] Step 4: Calculate the initial atmospheric transmittance based on the four-channel polarization intensity image;

[0178] Step 5: Calculate the target atmospheric transmittance parameters based on the atmospheric transmittance and the initial polarized atmospheric transmittance;

[0179] Step 6: Based on the target atmospheric transmittance parameter, calculate the atmospheric scattered light value, and based on the target atmospheric transmittance parameter and the atmospheric scattered light value, calculate the dehazed image;

[0180] Step 7: Calculate the target distance map based on the denoised photon echo data;

[0181] Step 8: Calculate the depth edge image based on the target distance map;

[0182] Step 9: Based on the depth edge image, perform detail enhancement processing on the dehazed image to obtain an enhanced dehazed image.

[0183] Corresponding to the dehazing image generation method provided in the embodiments of this application, the embodiments of this application also provide a dehazing image generation apparatus, such as... Figure 4 As shown, the dehazing image generation apparatus includes:

[0184] Acquisition module 401 is used to acquire raw photon echo data and four-channel polarization intensity images;

[0185] The filtering module 402 is used to perform noise filtering on the original photon echo data to obtain denoised photon echo data. The original photon echo data includes the photon count in each time window of each pixel.

[0186] The first calculation module 403 is used to calculate the target echo time of each pixel based on the denoised photon echo data.

[0187] The second calculation module 404 is used to calculate the target distance for each pixel based on the speed of light and the target echo time.

[0188] The third calculation module 405 is used to calculate the backscattering intensity map based on the denoised photon echo data, the target echo time of each pixel and the preset buffer interval.

[0189] The fourth calculation module 406 is used to calculate the target scattering coefficient based on the backscattering intensity map, the first calibration constant, and the target distance of each pixel;

[0190] The fifth calculation module 407 is used to calculate the atmospheric transmittance of each pixel based on the target scattering coefficient and the preset atmospheric scattering scaling factor.

[0191] The sixth calculation module 408 is used to calculate the initial atmospheric transmittance based on the four-channel polarization intensity image;

[0192] The seventh calculation module 409 is used to calculate the target atmospheric transmittance parameter for each pixel based on the atmospheric transmittance and the initial polarization atmospheric transmittance.

[0193] The eighth calculation module 410 is used to calculate the atmospheric scattered light value of each pixel based on the target atmospheric transmittance parameter.

[0194] The ninth calculation module 411 is used to calculate the dehazed image based on the target image, the target atmospheric transmittance parameter, and the atmospheric scattered light value.

[0195] Corresponding to the dehazing image generation method provided in the embodiments of this application, the embodiments of this application also provide an electronic device for performing the dehazing image generation method, such as... Figure 5 As shown, the electronic device includes: a processor 501; and a memory 502 for storing a program for generating a dehazed image. After the device is powered on and the processor runs the program for generating the dehazed image, it performs the following steps:

[0196] Acquire raw photon echo data and four-channel polarization intensity images;

[0197] The original photon echo data is subjected to noise filtering to obtain denoised photon echo data. The original photon echo data includes the photon count in each time window of each pixel.

[0198] Based on the denoised photon echo data, the target echo time of each pixel is calculated;

[0199] The target distance for each pixel is calculated based on the speed of light and the target echo time.

[0200] Based on the denoised photon echo data, the target echo time of each pixel, and the preset buffer interval, the backscattering intensity map is calculated.

[0201] The target scattering coefficient is calculated based on the backscattering intensity map, the first calibration constant, and the target distance for each pixel;

[0202] The atmospheric transmittance of each pixel is calculated based on the target scattering coefficient and the preset atmospheric scattering scaling factor.

[0203] The initial atmospheric transmittance was calculated based on the four-channel polarization intensity image.

[0204] Based on atmospheric transmittance and initial polarization atmospheric transmittance, the target atmospheric transmittance parameter for each pixel is calculated.

[0205] Based on the target atmospheric transmittance parameter, the atmospheric scattered light value of each pixel is calculated;

[0206] The dehazed image is calculated based on the target image, the target atmospheric transmittance parameter, and the atmospheric scattered light value.

[0207] Corresponding to the dehazing image generation method provided in the embodiments of this application, the embodiments of this application also provide a computer-readable storage medium storing a program for the dehazing image generation method, which is executed by a processor to perform the following steps:

[0208] Acquire raw photon echo data and four-channel polarization intensity images;

[0209] The original photon echo data is subjected to noise filtering to obtain denoised photon echo data. The original photon echo data includes the photon count in each time window of each pixel.

[0210] Based on the denoised photon echo data, the target echo time of each pixel is calculated;

[0211] The target distance for each pixel is calculated based on the speed of light and the target echo time.

[0212] Based on the denoised photon echo data, the target echo time of each pixel, and the preset buffer interval, the backscattering intensity map is calculated.

[0213] The target scattering coefficient is calculated based on the backscattering intensity map, the first calibration constant, and the target distance for each pixel;

[0214] The atmospheric transmittance of each pixel is calculated based on the target scattering coefficient and the preset atmospheric scattering scaling factor.

[0215] The initial atmospheric transmittance was calculated based on the four-channel polarization intensity image.

[0216] Based on atmospheric transmittance and initial polarization atmospheric transmittance, the target atmospheric transmittance parameter for each pixel is calculated.

[0217] Based on the target atmospheric transmittance parameter, the atmospheric scattered light value of each pixel is calculated;

[0218] The dehazed image is calculated based on the target image, the target atmospheric transmittance parameter, and the atmospheric scattered light value.

[0219] Corresponding to the dehazing image generation method provided in the embodiments of this application, the embodiments of this application also provide a computer program containing instructions, which, when executed by a computer, cause the computer to perform the following steps:

[0220] Acquire raw photon echo data and four-channel polarization intensity images;

[0221] The original photon echo data is subjected to noise filtering to obtain denoised photon echo data. The original photon echo data includes the photon count in each time window of each pixel.

[0222] Based on the denoised photon echo data, the target echo time of each pixel is calculated;

[0223] The target distance for each pixel is calculated based on the speed of light and the target echo time.

[0224] Based on the denoised photon echo data, the target echo time of each pixel, and the preset buffer interval, the backscattering intensity map is calculated.

[0225] The target scattering coefficient is calculated based on the backscattering intensity map, the first calibration constant, and the target distance for each pixel;

[0226] The atmospheric transmittance of each pixel is calculated based on the target scattering coefficient and the preset atmospheric scattering scaling factor.

[0227] The initial atmospheric transmittance was calculated based on the four-channel polarization intensity image.

[0228] Based on atmospheric transmittance and initial polarization atmospheric transmittance, the target atmospheric transmittance parameter for each pixel is calculated.

[0229] Based on the target atmospheric transmittance parameter, the atmospheric scattered light value of each pixel is calculated;

[0230] The dehazed image is calculated based on the target image, the target atmospheric transmittance parameter, and the atmospheric scattered light value.

[0231] It should be noted that for a detailed description of the dehazing image generation apparatus, electronic device, computer-readable storage medium, and computer program provided in the embodiments of this application, please refer to the relevant description of the embodiments of the dehazing image generation method provided in the embodiments of this application, which will not be repeated here.

[0232] Although this application discloses preferred embodiments as described above, it is not intended to limit this application. Any person skilled in the art can make possible changes and modifications without departing from the spirit and scope of this application. Therefore, the scope of protection of this application should be determined by the scope defined in the claims of this application.

[0233] In a typical configuration, an electronic device includes one or more processors (Central Processing Units), input / output interfaces, network interfaces, and memory.

[0234] Memory may include non-persistent storage in computer-readable media, such as random access memory and / or non-volatile memory, like read-only memory or flash memory. Memory is an example of computer-readable media.

[0235] Computer-readable media, including both permanent and non-permanent, removable and non-removable media, can store information using any method or technology. Information can be computer-readable operations, data structures, program modules, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PCM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EPROM), flash memory or other memory technologies, compact disc read-only memory (CD-ROM), digital video disc (DMCD) or other optical storage, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include non-transitory computer-readable media, such as modulated data signals and carrier waves.

[0236] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product implemented on one or more computer-usable storage media (including, but not limited to, disk storage, compact disc read-only memory, optical storage, etc.) containing computer-usable program code.

[0237] Although this application discloses preferred embodiments as described above, it is not intended to limit this application. Any person skilled in the art can make possible changes and modifications without departing from the spirit and scope of this application. Therefore, the scope of protection of this application should be determined by the scope defined in this application.

Claims

1. A method for generating dehazed images, characterized in that, include: Acquire raw photon echo data and four-channel polarization intensity images; The original photon echo data is subjected to noise filtering to obtain denoised photon echo data, wherein the original photon echo data includes the photon count in each time window of each pixel; Based on the denoised photon echo data, the target echo time of each pixel is calculated; The target distance for each pixel is calculated based on the speed of light and the target echo time. Based on the denoised photon echo data, the target echo time of each pixel, and the preset buffer interval, the backscattering intensity map is calculated. Based on the backscattering intensity map, the first calibration constant, and the target distance for each pixel, the target scattering coefficient is calculated. Based on the target scattering coefficient and the preset atmospheric scattering scaling factor, the atmospheric transmittance of each pixel is calculated. The initial atmospheric transmittance is calculated based on the four-channel polarization intensity image. Based on the atmospheric transmittance and the initial polarization atmospheric transmittance, the target atmospheric transmittance parameter for each pixel is calculated; Based on the target atmospheric transmittance parameter, the atmospheric scattered light value of each pixel is calculated; A defogging image is calculated based on the target image, the target atmospheric transmittance parameter, and the atmospheric scattered light value.

2. The dehazing image generation method according to claim 1, characterized in that, The calculation of the initial atmospheric transmittance based on the four-channel polarization intensity image includes: Stokes parameters are calculated based on the four-channel polarization intensity image, and the Stokes parameters include a first Stokes parameter, a second Stokes parameter, and a third Stokes parameter; Based on the Stokes parameters, the initial atmospheric transmittance for polarization is calculated.

3. The dehazing image generation method according to claim 2, characterized in that, The Stokes parameters calculated based on the four-channel polarization intensity image include: Based on the four-channel polarization intensity image, the Stokes parameters are calculated using the following formula: ； Among them, the Represents the first Stokes parameter, the The image representing the polarization intensity in the 0° polarization direction, the This represents a polarization intensity image along a 45° polarization direction. The image represents the polarization intensity along the 90° polarization direction. A polarization intensity image representing the polarization direction at 135°; ； Among them, the This represents the second Stokes parameter, the The image representing the polarization intensity in the 0° polarization direction, the A polarization intensity image representing the polarization direction at 90°; ； Among them, the This represents the third Stokes parameter, the This represents a polarization intensity image along a 45° polarization direction. This image represents the polarization intensity along the 135° polarization direction.

4. The dehazing image generation method according to claim 1, characterized in that, The calculation of the target atmospheric transmittance parameter for each pixel based on the atmospheric transmittance and the initial polarization atmospheric transmittance includes: Based on the atmospheric transmittance and the initial polarization atmospheric transmittance, the target atmospheric transmittance parameter for each pixel is calculated using the following formula: ； Among them, the The target atmospheric transmittance parameter represents the pixel in the i-th row and j-th column. Indicates adaptive weights, the The atmospheric transmittance is represented by the This represents the initial atmospheric transmittance at polarization.

5. The dehazing image generation method according to claim 4, characterized in that, Before calculating the target atmospheric transmittance parameter for each pixel based on the atmospheric transmittance and the initial polarization atmospheric transmittance, the method further includes: The target time window is determined based on the target echo time of each pixel and the preset buffer interval. The adaptive weights are calculated using the following formula: ； Among them, the For adaptive weights, the Represents the first signal-to-noise ratio, the Indicates the second signal-to-noise ratio; ； Among them, the Represents the first signal-to-noise ratio, the Indicates the first calibration coefficient. This represents the maximum photon count within the target time window; ； Among them, the Indicates the second signal-to-noise ratio, the Indicates the second calibration coefficient. Indicates median filtering. It is the gradient operator. Indicates to Find the gradient. This indicates the calculation of the Frobenius norm.

6. The dehazing image generation method according to claim 1, characterized in that, The calculation of the target echo time for each pixel based on the denoised photon echo data includes: ； Among them, the This represents the target echo time of the pixel in the i-th row and j-th column. This indicates finding the maximum value of the expression within the parentheses. This represents the photon count in the t-th time window of the i-th row and j-th column pixel in the denoised photon echo data. This represents the system's impulse response function.

7. The dehazing image generation method according to claim 6, characterized in that, The backscattering intensity map is calculated based on the denoised photon echo data, the target echo time of each pixel, and the preset buffer interval, including: Based on the denoised photon echo data, the target echo time of each pixel, and the preset buffer interval, the backscattering intensity map is calculated using the following formula: ； Among them, the This represents the backscattering intensity map of the pixel in the i-th row and j-th column. The target echo time of the i-th row and j-th pixel is represented by the following: Indicates the preset buffer interval, the This represents the photon count in the t-th time window of the i-th row and j-th column pixel in the denoised photon echo data.

8. The dehazing image generation method according to claim 7, characterized in that, The target scattering coefficient is calculated based on the backscattering intensity map, the first calibration constant, and the target distance for each pixel, including: Based on the backscattering intensity map, the first calibration constant, and the target distance for each pixel, the target scattering coefficient is calculated using the following formula: ； Among them, the The term represents the target scattering coefficient of the i-th row and j-th column pixel. Represents the first calibration constant, the This represents the backscattering intensity map of the pixel in the i-th row and j-th column. This represents the target distance of the pixel in the i-th row and j-th column.

9. The dehazing image generation method according to claim 8, characterized in that, The calculation of the atmospheric transmittance of each pixel based on the target scattering coefficient and a preset atmospheric scattering scaling factor includes: Based on the target scattering coefficient and the preset atmospheric scattering scaling factor, the atmospheric transmittance of each pixel is calculated using the following formula: ； Among them, the The atmospheric transmittance of the pixel in the i-th row and j-th column is described. The term represents the target scattering coefficient of the i-th row and j-th column pixel. This represents a preset atmospheric scattering scaling factor, the This represents the target distance of the pixel in the i-th row and j-th column.

10. A dehazing image generation apparatus, characterized in that, include: The acquisition module is used to acquire raw photon echo data and four-channel polarization intensity images; A filtering module is used to perform noise filtering on the original photon echo data to obtain denoised photon echo data, wherein the original photon echo data includes the photon count in each time window of each pixel; The first calculation module is used to calculate the target echo time of each pixel based on the denoised photon echo data. The second calculation module is used to calculate the target distance for each pixel based on the speed of light and the target echo time. The third calculation module is used to calculate the backscattering intensity map based on the denoised photon echo data, the target echo time of each pixel, and the preset buffer interval. The fourth calculation module is used to calculate the target scattering coefficient based on the backscattering intensity map, the first calibration constant, and the target distance of each pixel; The fifth calculation module is used to calculate the atmospheric transmittance of each pixel based on the target scattering coefficient and the preset atmospheric scattering scaling factor. The sixth calculation module is used to calculate the initial atmospheric transmittance based on the four-channel polarization intensity image; The seventh calculation module is used to calculate the target atmospheric transmittance parameter for each pixel based on the atmospheric transmittance and the initial polarization atmospheric transmittance. The eighth calculation module is used to calculate the atmospheric scattered light value of each pixel based on the target atmospheric transmittance parameter. The ninth calculation module is used to calculate the dehazed image based on the target image, the target atmospheric transmittance parameter, and the atmospheric scattered light value.

Images